Commercial Microgrids: The Next Big Thing?
The trouble that utilities have had maintaining service during hurricanes and other major storms has led to a renewed interest in microgrids -- small communities that can generate their own electricity and avoid blackouts during periods when the grid is down as well as save money by buying power during off-peak hours while generating their own electricity during peak hours. How widespread are microgrids? How great a threat are they to traditional utilities? A group of panelists talked about these and other issues at an Infocast conference on commercial microgrids in Washington in late April.
The panelists are Mark Crowdis, president of Think Energy Inc., Michael Kornitas, energy conservation manager for Rutgers University, Brian Patterson, chairman of Emerge Alliance, Jeff Seidel, director of capital expenditures for the Mohegan Tribal Gaming Community Authority, Dr. Mohammad Shahidehpour, professor and director of the Robert W. Galvin Center for Electricity Innovation at the Illinois Institute of Technology, and Phil Smith, director of federal project development for Honeywell Building Solutions. The moderator is Keith Martin with Chadbourne in Washington.
MR. MARTIN: Is there a difference between a "microgrid" and just generating your own power? Paper companies used to generate their own power.
DR. SHAHIDEHPOUR: Microgrids are nothing new. If you go to most villages in the third world, they have a diesel generator and it is not connected to any state grid. Microgrids were in existence for many years, but now people are using smart systems to manage the load more intelligently. Microgrids are different because now there are data and control systems that make microgrids a lot more efficient.
MR. MARTIN: Will we see more communities in the United States with their own grids and their own power supplies?
MR. CROWDIS: I have a client in Hawaii who is hopping mad with the local utility. It is entirely possible for small communities to disconnect from the grid in the higher-cost markets.
MR. MARTIN: Will those communities be regulated as utilities?
MR. CROWDIS: They would have to be. However, there is a size issue. One of our clients is Anguilla, which has a 10-megawatt average demand and 15-megawatt peak demand. On a small island like Anguilla, a community disconnecting from the grid would have too small a load to make it economic.
MR. KORNITAS: We are more likely to see a move to distributed generation on a massive scale that will give rise to individual homeowners as generators. It will follow the model of the internet. There will be some large entities and some small entities.
MR. PATTERSON: Distributed generation and the internet have the same characteristics. Your laptop manages its own power and can be on the grid or off the grid. It can run your USB light and all kinds of other things you can plug into it. The only thing that prevents you from doing that today with power is regulations about crossing rights of way and things like that. Without such regulations, when the solar system on your roof is not being used fully, you could sell the spare electricity to your neighbor.
DR. SHAHIDEHPOUR: We struggle with that issue in the Chicago area. The reason universities can do fancy things is because we own the grid. Everything is behind the meter. About 60 communities in the Chicago area over the last year have formed community choice operations. The struggle is that they do not own the grid. If they want to connect any item to the system, they have to have the utility's permission. That permission is going to be a very long time coming. They try to operate as virtual microgrids managing their loads without controlling the electrons, but it will be an uphill battle.
MR. MARTIN: What is a virtual microgrid?
DR. SHAHIDEHPOUR: A virtual microgrid is basically where you do not own the generation or the grid. You only own the load, and you manage the load.
A true microgrid must be able to be operated as an island. You have to own the generation and these entities do own generation, but in order to get the power to the load, they have to go through wires that they do not own. In many cases, the utility is in favor of letting them manage the wires.
MR. PATTERSON: That model is going to change because you have a lot of people who can climb utility poles today and are already well equipped to move low and even medium voltages. It all comes down to policy. We saw the same thing happen with telephones when the publicly-granted monopolies existed, and the incumbent telephone companies said you cannot use our telephone lines and you have to operate under our rules. There was a battle between the Federal Trade Commission and the Interstate Commerce Commission, and the Federal Communications Commission broke that deadlock. We might see a similar battle soon.
MR. KORNITAS: Even if a community is an island, it would still need to be connected to the grid in case the microgrid's power goes out so that it can pull power from somewhere else to keep equipment running.
MR. MARTIN: I want to establish how many different models of microgrid there are. Princeton and Rutgers operate basically as islands and generate electricity for themselves. What other models are there?
DR. SHAHIDEHPOUR: Microgrids are either connected to the grid or not. If a microgrid is in the middle of nowhere, then it is obviously operated as an island. If it is in a metropolitan area, then it is often connected to the grid, even though it is run as an island. In our university's case, we only run our microgrid as an island for liability reasons. We do not operate the system as an island for economic reasons.
MR. MARTIN: Is there any place in the United States where a microgrid can operate today involving an entire community without being a municipal utility?
MR. PATTERSON: There are several if I am not mistaken. They are identified in the book Perfect Power. There are other experiments that are being managed by the Electric Power Research Institute that involve more than one building or campus.
Northern Power Systems created a system that uses wind, solar and diesel at the North Pole for a weather station. That little system ran on its own. To me, a microgrid is something that can work independently of the utility grid.
In most cases, you are not going to be able to connect to the grid as a whole community unless you are a municipal utility.
MR. MARTIN: Is anyone aware of a movement to try to relax the regulatory restrictions in any states to allow broader microgrids?
DR. SHAHIDEHPOUR: It is not going to happen soon because the appeal of the business model is too limited. You have to convince utilities that there is something in it for them to promote microgrids. Some utilities are proactive, but overall it will be difficult to convince utilities to facilitate taking a chunk of the load away from them.
MR. CROWDIS: Utilities might find microgrids appealing in cases where the utilities would otherwise have to make large repairs or upgrades to their transmission systems. A utility might embrace a microgrid if it can avoid having to invest additional dollars in the transmission system.
DR. SHAHIDEHPOUR: Many businesses in the Chicago area have backup power. If everybody has a backup generator, the marginal cost of operating that system is significantly more expensive than starting a microgrid where you have a coordinated way of controlling the electricity flow in the region. At some point utilities are going to see that it is to everyone’s benefit to have microgrids, but it will be a while.
MR. CROWDIS: Economics will drive policy. There will come a day when utilities in the US will conclude it is uneconomic to serve certain areas. They will realize that re-stringing power lines every other year in problem areas is a costly pain so a microgrid might be better.
MR. MARTIN: Jeff Seidel, do you already generate your own power for the Mohegan casinos?
MR. SEIDEL: We generate some electricity, but not a lot. We are looking at generating more. The issue is whether we can produce the power more cheaply then we can buy it. Our load maxes out at about 28 megawatts in the summer for our largest casino. We can only generate about eight megawatts from back-up generation and fuel cells today. We want to expand, but it has to make sense with the heating systems. If we expand, it will probably be with more cogeneration.
MR. MARTIN: Michael Kornitas, the load at Rutgers University is more than 20 megawatts?
MR. KORNITAS: We have a 13-megawatt cogeneration plant, and we just added a 1.4-megawatt solar facility and an 8.8-megawatt solar array that is over our parking lots. We produce a lot of our power, but we also buy power to cover the load. The solar covers 65% of one of our campuses and the cogeneration facility covers quite a bit of the other. We dispatch our energy based on economics. Sometimes it is cheaper for us to buy electricity from the grid than to generate it.
MR. MARTIN: We heard immediately before this panel that Princeton University is managing its purchases from the grid not only based on price but also based on its environmental goals. It tries to manage its power supplies to limit the amount carbon dioxide its electricity usage is responsible for emitting. Is Rutgers doing the same thing?
MR. KORNITAS: Yes. We have a memorandum of understanding with the US Environmental Protection Agency under which we agreed to send EPA our greenhouse gas data in six-month intervals. We monitor the effect our electricity usage has on emissions closely.
MR. MARTIN: Are you thinking of adding more generating capacity?
MR. KORNITAS: Yes. We are looking at this currently, but the key is whether it makes economic sense. We have to look at both environmental and economic factors. We are looking now at a natural gas fuel cell which is basically cogeneration on a different level. The fuel cell would not have variable output, but it could be part of the base generation.
MR. MARTIN: Do you own the solar systems on your campus? Or does a solar company own them and lease them to you or just sell you electricity under a long-term power purchase agreement?
MR. KORNITAS: We own the 1.4-megawatt solar facility. We lease the 8-megawatt solar array and the amount we get for the solar renewable energy credits covers the rent to the lessor.
MR. MARTIN: The SRECS cover the full rent despite the fact that they fluctuate in value over time? What are SRECs worth today in New Jersey?
MR. KORNITAS: About $110 a mWh.
MR. MARTIN: $110 now, but the long-term outlook is a little murky, right?
MR. KORNITAS: Yes, but we were very conservative with our numbers, so we feel we will be okay for the life of the facility.
MR. MARTIN: Dr. Shahidehpour, what is the total load for the Illinois Institute of Technology?
DR. SHAHIDEHPOUR: The total load is about 12 megawatts. The total generation is about 10 megawatts. We buy some power, and we do demand response and load control. Eight megawatts of the generation is gas, and the rest is from a mix of solar, batteries and wind.
We are looking into adding more storage. One issue is that in northern Illinois, the price of electricity at night is negative because of the local nuclear facilities. So it makes economic sense to have batteries available. You basically get paid to charge the batteries and then you get paid to discharge them. Right now we have a ZBB or zinc bromine battery on campus that we use to provide frequency regulation services to the PJM grid.
MR. MARTIN: So you provide ancillary services to the grid and you earn money.
DR. SHAHIDEHPOUR: We are looking at other generation options as well, like geothermal. There is a good chance we will add geothermal, but the option in which I am most interested is batteries.
MR. MARTIN: Philip Smith, I think of Honeywell as a contractor. Do you also generate your own electricity on some of your campuses?
MR. SMITH: Not too often. I mentioned earlier a project in which are involved at the US Food and Drug Administration headquarters. Honeywell installed a cogeneration unit on the FDA campus and is paid a percentage of the energy savings each year under a long-term energy savings performance contract, which is our core business. It was a new campus development, and we were able to finance construction of the CHP plant by borrowing against the future energy savings performance payments.
We went from what was originally planned to be a thermal-only central plant to a cogeneration plant that distributes hot water for comfort heating. There is still a conventional chill water plant, although we introduced some absorption and chillers into the mix and then we provided for the backup power. The original driver was power reliability. We run in parallel with the grid. We are generally in a net export mode with the grid, but we do thermal load following in the economic model. We are now an energy-only resource on the PJM grid, so we get locational marginal pricing for what we put on the grid and it shows up as a credit on the federal government's overall utility bill for the area.
Relationship with the Grid
MR. MARTIN: Why would it make sense to disconnect from the grid? I did not hear any of you say you are disconnecting from the grid. Is anybody backing off the grid entirely?
DR. SHAHIDEHPOUR: We do it only for reliability reasons. If the area is in an outage, we can keep our lights on by disconnecting the campus and turning it on as a microgrid. We have divided the campus into seven loops, and each loop is connected to one of the substations. Each loop is equipped with S&C Vista switches that allow the campus to disconnect almost instantly after detecting a fault in power supply. Ever since we have done that, we have had no downtime on campus.
MR. MARTIN: Michael Kornitas, does Rutgers ever supply excess electricity to the grid?
MR. KORNITAS: No, we do not. We use all the electricity we produce on campus.
MR. PATTERSON: A microgrid allows you to articulate power in a different way than when you are connected or having to use the common denominator of an interconnected grid. To the extent that you are disconnected or isolated, you can set different quality parameters for the power you generate or store. For example, data centers require a different quality of power, and that is one of the reasons that data centers are largely isolated from the line-voltage power.
MR. MARTIN: In terms of opportunity in this sector, I heard there is an opportunity to add more storage and more generating capacity in the form of solar, fuel cells, geothermal heat pumps and CHP or cogeneration facilities. Are these the main opportunities in this sector?
MR. CROWDIS: I think solar, storage and wind are opportunities. We worked for about six years on microwind projects. I am not completely convinced that those things can work, but large wind projects on site will become more attractive over time if they can be combined with storage.
When we talk about microgrids here in the United States, disconnecting from the grid does not make sense because the utility is usually of high quality and the power supply is reliable. In certain markets, such as the Dominican Republic, it makes much more sense to have control over your own power supply.
MR. MARTIN: So control over your own power supply makes less sense in the United States?
MR. CROWDIS: There are two places where there might be power supply issues, making a microgrid practical. One is Alaska. There are very remote communities there. Some people have said that they can achieve payback with a solar photovoltaic array very quickly even with no sun for half the year. With the cost of storage coming down, some remote communities may do that. We also might see it in Hawaii.
MR. SMITH: We are focused primarily on reliability. We look at federal entities that have critical loads. We look at ways to serve them through some sort of viable economic model.
We do not separate from the grid. We run in parallel with the grid so that we have enhanced reliability, but during the course of the last couple years, we have disconnected 50 times. Some of that was proactive, and some was automatic. In the case of Hurricane Irene, we saw the storm coming so we went into island mode. In the case of the earthquake, we did not anticipate that so we disconnected automatically.
MR. MARTIN: Michael Kornitas, you must be pitched by cogenerators, solar companies and storage purveyors. Who else?
MR. KORNITAS: Fuel cells seem to be the big thing because of the state subsidies. We are looking at storage right now. It would be the greatest thing in the world if we could get storage at a price that makes sense to install it. We are looking at thermal storage right now.
MR. CROWDIS: We released a request for proposals, and we will get bids this week for a five-star hotel in the Virgin Islands. We have been contacted by virtually every company on the planet that is doing some kind of storage indoor microgrid work. Some of the pricing is very attractive.
MR. MARTIN: How much does storage cost per megawatt of capacity?
MR. CROWDIS: It depends on how big the application is. We had one deal that was three mWhs of storage for around $3 million. The deal was paired with renewables.
MR. MARTIN: You could also earn revenue from ancillary services. What percentage of the cost do you think can be covered by providing services to the grid?
MR. CROWDIS: None of the deals at which I am looking would provide that service.
MR. MARTIN: Jeff Seidel, who is pitching you at the moment?
MR. SEIDEL: Mostly solar and cogeneration companies.
MR. MARTIN: Brian Patterson, where do you see opportunities?
MR. PATTERSON: I do not think the goal is to be disconnected from the national grid. Microgrids are more about providing different levels of quality and reliability. In essence, they make the user share responsibility with the utility. The goal is not to replace the utility.
Shift in Basic Utility Model?
MR. MARTIN: Dr. Shahidehpour, do you think we are on the verge of a change in how electricity is supplied in this country? Should the regulated utilities be worried?
DR. SHAHIDEHPOUR: We are adding so much uncertainty to the electric power system today with intermittent sources of electricity like solar and wind. Electric cars are another element. The system is volatile. It is very difficult to assume one control center for a large utility where we would be able to manage all these uncertainties within the system. By localizing the control, we would improve the operation of the system from an economic and a reliability point of view.
It will be a while before we have system partitioning, but something has to happen because utilities do not have the answer to solar and wind adding uncertainty to projections of electricity output. By localizing, we are still connected to the grid but we make their lives much more manageable.
MR. CROWDIS: My view is that microgrids are only being pushed by the US military, islands and remote communities because they need the service. India and China are doing all kinds of things. I feel there will be more of a boomerang effect.
Let’s look at the 900-pound gorilla. Natural gas is cheap. It will remain cheap for a while. I do not see a mass migration to microgrids. I do not think see it happening in the continental United States. We have great utility service at low prices.
AUDIENCE MEMBER: There has to be some sort of new regulatory compact to allow the utilities to move toward this different model. The utilities have huge investments in wires and generating facilities that they are recovering over time through rates. To the extent some customers disconnect from the utility grid, then the remaining customers are left having to the bear the stranded cost of a system that was built to serve a larger customer base. That does not work.
MR. CROWDIS: A hotel in Antigua decided recently to take its large reverse osmosis system off the grid and run it with solar. I think this idea that you can take things off the grid and run them independently is disruptive. What I am worried about in Antigua is that it is an impoverished place, and what happens to the rest of the ratepayers? I am concerned about equity and fairness.
MR. MARTIN: Is it true that if you install a rooftop solar system, you will be protecting yourself from utility outages? Can such a system operate when the grid is down?
MR. KORNITAS: We would have to shut it down. When the grid goes down, we are not allowed to produce solar.
MR. SMITH: The amount of PV that we have on the Food and Drug Administration campus is about 30 kilowatts out of 26 megawatts of capacity, so it is a tiny part of the overall generating capacity. Solar has played a critical role in Japan in the post-tsunami period. It plays a critical role when it is available in the post-restorative period because it can bridge a gap.
MR. CROWDIS: I have been talking to a number of investors who are getting comfortable with the idea of large microgrids in which they might invest. We are at the same stage on the learning curve as in the early days of distributed solar.